Member in the Spotlight: Le An

In this series, we introduce a member of our Society by means of an interview. This month, we are happy to introduce Le An from University of California, Riverside, who recently finished his PhD thesis, "Real-World Person Identification”.

What are your research interests in the signal processing field?

I started my research in signal processing, specifically in the area of telecommunication, when I was doing undergraduate research projects. Later on I found myself more interested in image processing where I could actually see the impact of signal processing vividly in color or gray images. As I started my PhD study, I have been working on image processing algorithms that can aid other areas such as computer vision and pattern recognition.

Could you briefly introduce your research?

My research focuses on person identification in real-world. For example, I have developed a face super-resolution algorithm which can improve the face recognition accuracy when low-resolution faces are to be recognized. I have recently been working on person re-identification, the goal of which is to match individuals across different cameras using their body signatures. In addition I am also working on emotion understanding through facial expression recognition.

In your opinion, what was the most impressive result published in IEEE SPS journals and conferences within the last 12 months?

I feel that an impressive result in publication should enable or enhance a lot of real-world applications. I think a recent paper entitled “Image Quality Assessment for Fake Biometric Detection: Application to Iris, Fingerprint, and Face Recognition” published in IEEE Transactions on Image Processing by Galbally et al. provides a very useful fake detection method that can work on different biometrics and this technique can enhance many security applications.

Could you introduce an important state-of-the-art research issue (or technology) in this field (Other than your research)?

With the recent surge of big data, deep learning has pushed the frontiers of many research areas such as speech recognition and image classification. I believe that as more data become available, large scale deep learning using big data will gain more interests and will have a greater impact on many aspects of our lives. Artificial intelligence may develop rapidly with abundant data and the fast growing capability of computers.

In which way have you been connected first with IEEE SPS?

The first connection with IEEE SPS was my early PhD work submitted to IEEE International Conference on Image Processing. Since then I have been actively submitting my work to IEEE SPS journals and conferences such as IEEE Transactions on Information Forensics and Security, IEEE Signal Processing letters, IEEE International Conference on Image Processing, etc.

In which way did you know the IEEE SPS e-NewsLetter?

I have been receiving the IEEE SPS e-NewsLetter and it has been a great source for me to get familiar with the cutting-edge signal processing technologies in different areas.